Major's 50% and 80% lists

Stephen Hughes
Posts: 3323
Joined: February 26th, 2013, 7:12 am

Major's 50% and 80% lists

Post by Stephen Hughes » November 7th, 2015, 1:12 am

In his October 30th article he notes:
The 50% list consists of just 27 lemmata. The only verbs are γίνομαι, εἰμί, ἔχω, and λέγω. The only nouns are θεός, κύριος, and Ἰησοῦς.
That is interesting to me because the first two of those verbs are for expression in the language, not meaning and ἔχω is also sometimes just for expression. The three nouns are the characters of the narrative and and objects of description and discussion, and λέγω is a narrative verb.

Two things I deduce from that. Firstly, mastering the grammatical words of the language is a universal need. So to count vocabulary, these grammatical words can first be excluded then frequency for other meaningful words could be tallied for sub-sections. Secondly, it is a clear that nouns are themselves are outside the core (grammatical vocabulary), and need to be learnt differently.
0 x


Γελᾷ δ' ὁ μωρός, κἄν τι μὴ γέλοιον ᾖ
(Menander, Γνῶμαι μονόστιχοι 108)

Paul-Nitz
Posts: 428
Joined: June 1st, 2011, 4:19 am
Location: Lilongwe, Malawi

Re: James Tauber is blogging daily until SBL

Post by Paul-Nitz » November 15th, 2015, 6:48 am

I've done quite a bit of playing with Major's 50% and 80% lists, attempting to reduce it to suit both NT Greek and communicative teaching. The latter requires some words that would be typically used in classroom work. I'm thrilled to see that Tauber analyzed the GNT this way. He offers a link to the code in his article, but I haven't slightest idea how to make that practically useful. How can I get a simple list of the lemmas and the forms that would be included in Tauber's 50% and 80% list?
0 x
Paul D. Nitz - Lilongwe Malawi

Stephen Hughes
Posts: 3323
Joined: February 26th, 2013, 7:12 am

Re: James Tauber is blogging daily until SBL

Post by Stephen Hughes » November 16th, 2015, 2:59 pm

Paul-Nitz wrote:playing with Major's 50% and 80% lists
Well, taking a glance at them just now, if you are adapting them, not just using them outright, here are a few points you may wish to weigh up for yourself:
εἰκός, εἰκότος, τό proper, probable
The substantive might perhaps be "likelihood", "probability".
στερέω separate
There may be other words that could be used here, like "rob".
ἔλεγος –ου, ὁ a lament
This is not the most common meaning. Perhaps "melody".
0 x
Γελᾷ δ' ὁ μωρός, κἄν τι μὴ γέλοιον ᾖ
(Menander, Γνῶμαι μονόστιχοι 108)

jtauber
Posts: 60
Joined: May 31st, 2011, 11:34 am
Location: Burlington, MA, USA
Contact:

Re: James Tauber is blogging daily until SBL

Post by jtauber » November 16th, 2015, 4:22 pm

Paul-Nitz wrote:I've done quite a bit of playing with Major's 50% and 80% lists, attempting to reduce it to suit both NT Greek and communicative teaching. The latter requires some words that would be typically used in classroom work. I'm thrilled to see that Tauber analyzed the GNT this way. He offers a link to the code in his article, but I haven't slightest idea how to make that practically useful. How can I get a simple list of the lemmas and the forms that would be included in Tauber's 50% and 80% list?
Let me generate the full lists and link to them here.
0 x
James Tauber
http://jktauber.com/

jtauber
Posts: 60
Joined: May 31st, 2011, 11:34 am
Location: Burlington, MA, USA
Contact:

Re: James Tauber is blogging daily until SBL

Post by jtauber » November 17th, 2015, 12:22 am

jtauber wrote:
Paul-Nitz wrote:I've done quite a bit of playing with Major's 50% and 80% lists, attempting to reduce it to suit both NT Greek and communicative teaching. The latter requires some words that would be typically used in classroom work. I'm thrilled to see that Tauber analyzed the GNT this way. He offers a link to the code in his article, but I haven't slightest idea how to make that practically useful. How can I get a simple list of the lemmas and the forms that would be included in Tauber's 50% and 80% list?
Let me generate the full lists and link to them here.
Okay, I've made the GNT 50% and 80% lists available at https://github.com/jtauber/core-gnt-vocab along with Dodson's glosses.

You can get the 50% list directly at https://raw.githubusercontent.com/jtaub ... mma_50.txt
and the 80% list directly at https://raw.githubusercontent.com/jtaub ... mma_80.txt

If you have any suggestions on improving the glosses, file issues or pull requests (if you know how) or mention them here.

I'm also interested if people think certain lexeme should be split like Major does (e.g. suppletive verbs).
0 x
James Tauber
http://jktauber.com/

Stephen Hughes
Posts: 3323
Joined: February 26th, 2013, 7:12 am

Re: Major's 50% and 80% lists

Post by Stephen Hughes » November 20th, 2015, 3:23 pm

I've mentioned a few times about my basic mistrust of frequency lists. Let me discuss that a little here and suggest a couple of possible ways to modify them to make them more useful.

One problem is polysemy. Simply counting words gives the frequency of words in the totality of all their meanings. There are two (or three - being the neither of the two) ways that glosses might be chosen or arranged; perhaps etymologically (more ancient or less developed first), or by the frequency of which meanings are most common. I strongly suspect, that the "easiest first" ethos would tend to arrange the meanings with the least developed first. The effect of that is to require students (of the Bible) to (unguidedly) trace through a series of steps in the logical development and contextualisation of the meaning of words, both of which require skills that they have neither been trained in nor given any guidance in. The end result is that the semi-learned come up with "the real meaning" statements about the Greek. I think that if the second method of arrangement were to be favoured, the 50% frequency lists would remain about the same, but the 80% would also come to contain a higher proportion of grammatical words - the words that primarily reflect the structure of the language.

The second problem is grammatical words. The implication of having a group of extremely high frequency words is that there are a group of the words of Greek that are used in all situations and texts of the Greek language - the words that make Greek Greek. I don't think it is sound to learn those words by means of glosses. They do not have different "meanings", but rather different syntactic functions. They only really make sense within syntactic and discourse patterns. They need to be learnt in conjunction with the acquisition of grammar. Seeing as they are roughly speaking ubiquitous, they needn't be included in frequency tables. That is to say, someone wanting to read the language needs to know the language - those grammatical / syntactic words. The order of introduction needs to be tailored to the way Greek syntax builds and combines to form the language. They are all necessary to learn, even if they fall some distance outside the highest frequencies.

The frequency of the other words in the language (besides the grammar words) can then be counted. What words among them that are frequent are what is going to make a particular text readable. Very roughly, subtracting most of the 50% list from the 80% list to give a list of non-grammatical words - ones that could to some greater extent be beneficial to learn by glosses, but what about the meanings?

Should words be learnt on the basis of their being an identifiable form of the word, or because they are a particular form with a meaning. If the assumption in preparing people to interact with the text is that they will encounter a form, decide from a list of possibilities which one is most suitable for translating it a particular passage, then perhaps arranged by forms is okay. But in that way, there is always a lot of processing to be going on. Counting in that way is like counting the verbs "charge (as in mobile phone)" together with "charge (as in ask for monetary payment)", "charge (as in lay charges)", "charge (as in run headlong into)", and "charge (as in give a responsibility)" and saying that their combined frequency justifies the word (and all its meanings) being learnt at a certain stage of learning. Of course, we know intuitively which ones are more frequent. In this case the first three (individually quite high in frequency in their own fields - mobile phone usage, commerce and law) combine to make the word (form) frequent and therefore make the infrequently used meanings seem much more frequent than they are. A NTG case in similar point would be λόγος. The first group of meanings (as listed in BDAG) is the "speech act" meanings, what is termed "communication whereby the mind finds expression". By far and away that is the largest group. Then there are the "reckoning" meanings - limited to accounting contexts. Then there is the meaning that is only used in (as I count from the entry) 4 instances "Logos" what they as express as "the independent personified expression of God". In the semi-learned popular imagination just about any instance of λόγος could be the "Logos". The theory (almost a conspiracy theory) that Logos is lurking in the shadows of the meaning of every λόγος has a rather strong following.

Beyond sensationalistic examples now, a realistic representation of words by frequency in conjunction with sub-frequencies by gloss, or counting frequencies by form and gloss, rather than just gloss, really makes more sense than counting just by form.

The obvious problem is that there doesn't seem to be open source data to support such counting. I've looked about for even Thayer's in an open-source digital version, but even biblehub.com uses their presentation of Thayers by permission of biblesoft.com. Of course neither Thayer's, nor even BDAG contain the information where which word is allocated - or have a consensus on to which meaning each usage should be allocated to. I think that in that case, while the overall total frequencies can not go beyond the total number of occurrences in the text, the frequency sub-totals could add up to be more than the overall totals. Where a disputed or uncertain could be marked. Eventually, the figures for sub-frequencies by gloss would improve, but I don't think that disputes by scholars need to be normalised just for the sake of having numbers add up.

I note that the asterisks that were in BAGD are no longer in BDAG.

Does anyone have suggestions how any of that could be done?
0 x
Γελᾷ δ' ὁ μωρός, κἄν τι μὴ γέλοιον ᾖ
(Menander, Γνῶμαι μονόστιχοι 108)

Paul-Nitz
Posts: 428
Joined: June 1st, 2011, 4:19 am
Location: Lilongwe, Malawi

Re: Major's 50% and 80% lists

Post by Paul-Nitz » November 21st, 2015, 10:14 am

James, Here's my list:
https://dl.dropboxusercontent.com/u/98867968/CORE COMMUNICATIVE GREEK Vocabulary.xlsx\

The purpose of this list is to serve as the pool of words I try to restrict myself to in my beginning koine Greek classes. I am a big believer in teaching structures and only as much vocabulary as is needed to learn structures. Vocabulary is quickly and easily added through reading.

Creating this list, I manually looked through Major, compared NT frequency lists, made some arbitrary decisions, added some non-NT words.
0 x
Paul D. Nitz - Lilongwe Malawi

jtauber
Posts: 60
Joined: May 31st, 2011, 11:34 am
Location: Burlington, MA, USA
Contact:

Re: Major's 50% and 80% lists

Post by jtauber » November 22nd, 2015, 8:52 am

I should note that, even though I produced the lists, I also am:

(a) not a fan of purely frequency-based approaches
(b) a big believer in showing words in constructions rather than glossing them in isolation

While I would love a full-blown pattern dictionary / construct-i-con, one less ambitious thing I'd like to see developed (and would love to help facilitate) is a list of exemplar phrases / clauses (whatever is appropriate for the class of the given word) for each common word, potentially organised by sense.

James
0 x
James Tauber
http://jktauber.com/

Shirley Rollinson
Posts: 302
Joined: June 4th, 2011, 6:19 pm
Location: New Mexico
Contact:

Re: Major's 50% and 80% lists

Post by Shirley Rollinson » November 22nd, 2015, 6:52 pm

Paul-Nitz wrote: - - - snip snip - - -
. I am a big believer in teaching structures and only as much vocabulary as is needed to learn structures. Vocabulary is quickly and easily added through reading.
- - snip snip - - - .
But when will they read (at first) except in or for class ? And how will they read easily, rather than painfully, if they don't have a working vocabulary?
When learning a language, I've always found that lack of vocabulary is one of the things that causes despondency and the temptation to give up. I can learn all the fancy grammar and chant the paradigms, but if I can't ask for "a gallon of gas and a dozen eggs, and how much does it cost?" (or the GNT equivalent) then the language becomes a series of hurdles rather than a means of communication.
Shirley Rollinson
0 x

Paul-Nitz
Posts: 428
Joined: June 1st, 2011, 4:19 am
Location: Lilongwe, Malawi

Re: Major's 50% and 80% lists

Post by Paul-Nitz » November 23rd, 2015, 8:00 am

Shirley Rollinson wrote:And how will they read easily, rather than painfully, if they don't have a working vocabulary?
I wrote a related blog some time ago:
"Flush Cards for Greek Vocabulary"
https://www.linkedin.com/pulse/20140908 ... vocabulary
0 x
Paul D. Nitz - Lilongwe Malawi

Post Reply